Paul Brest and Kelly Born recently published a fascinating article in Stanford Social Innovation Review called “When Can Impact Investing Create Real Impact?”   In additional to this article being very much worth the read as a major contribution to the theoretical underpinnings of impact investing, SSIR took the step of framing the piece as “Up for Debate” and they asked a number of leading thinkers in the space to share their perspectives on the article.

The combination of the original article and the responses serves as one of the most up-to-date, unvarnished perspectives on where impact investing is today, what different leading players see as the strengths and fault lines within the space, and where people come out on the core questions of what is (and is not) investing for impact, how social impact and financial returns are either complementary or in tension, and the role of measurement (assessing impact) in creating long-term change.

For those of you who are in impact investing or want to understand the space, I highly recommend the full suite of responses by Audrey Choi (Morgan Stanley), Nancy Pfund (DBL Investors), Amit Bouri (GIIN), Beth Richardson (B Lab), Brian Trelstad (Bridges), Mike McCreless and Catherine Gill (Root Capital), Matt Bannick and Paula Goldman (Omidyar), Sterling Speirn (Kellogg Foundation),  Nick O’Donohoe (Big Society Capital), Antony Bugg-Levine (Nonprofit Finance Fund), John Goldstein (Imprint Capital), Harold Rosen (Grassroots Business Fund), David Wood (Hauser Center), Alvaro Rodriguez and Michael Chu (IGNIA),

My own take on the article was this:

Paul Brest and Kelly Born’s article brings a welcome analytical framing to the emerging field of impact investing. The sector has been growing exponentially, with a large number of new funds being raised and increasingly mainstream visibility. In June 2013, UK prime minister David Cameron kicked off a G8 session in Lough Erne, Northern Ireland, on impact investing.

The irony is that despite the increased attention and funding for impact investing, there is still considerable debate about what actually constitutes an impact investment. Indeed, some of the most aggressive claims about the size of the impact-investing market—generally agreed to be a few billion dollars today, and predicted by some to grow to anywhere between a few hundred billion and $1 trillion by 2020—hinge on loose definitions. For example, much of the predicted hundreds of billions in impact investing comes from the microfinance sector, capital that effectively has been rebranded “impact investment.” And a considerable amount of private capital that was already being invested in developing countries seems increasingly to be called impact investments.

This lack of clarity about what impact investing is and isn’t makes it harder for investors to understand the landscape, harder for funds to raise money, and harder for entrepreneurs to navigate a sea of new investors describing themselves in similar ways but behaving very differently in terms of return expectations, risk appetites, and long-term objectives.

Brest and Born’s article bravely takes on many of these core definitional issues, and it is meticulous in defining the types of impact an investor could have: enterprise (the product or service makes a difference), investment (provision of capital that would otherwise not be available or be more costly), and nonmonetary (supporting the ecosystem) impacts.

Nonmonetary impacts are the clearest and least controversial—providing technical assistance, building the enabling environment or the ecosystem, even bringing in more mainstream capital. Most of these activities are familiar forms of enterprise support (though grossly underdeveloped within impact investing). Investment impact, in Brest and Born’s view, hinges on the concept of additionality—to be an impact investor, one must be making something happen that otherwise wouldn’t. For example, one can be an impact investor seeking high rates of financial return if one is exploiting frictions in the market (such as small deal size or misperception of trade-offs between risk and returns) that keep others from deploying capital. Appropriately, return expectations alone tell you little about who is in and who is out.

In effect, this definition narrows the field of impact investing. By asserting that an investor in a highly developed subsector of the market—where people compete for deals and capital flows freely—is not an impact investor seems to imply, for example, that much of the capital going in to microfinance today is not impact investment. This is an interesting assertion (nothing happened that wouldn’t have otherwise) but it also feels like analytical parsing: it is true that an investor is not having impact if she is doing something that would happen anyway, but that tells you little about whether the investment has created enterprise impact, which seems like a more important question.

This brings us to the central question of enterprise impact. Ultimately, what we care about is whether, how, and why impact investments improve peoples’ lives. Yet unfortunately, besides a broad framing, the article does not dig deep into enterprise impact outside of the wholly accurate, if nearly clinical, observation that “the absence of data and analysis makes it difficult for impact investors to assess the social impact of the enterprises they invest in.”

As such, the article reflects the current state of dialogue in impact investing, which I hope marks the closing of the first chapter of sector-formation and the start of the second. Nearly all of the discussion in impact investing currently is focused on the capital-formation end of the value chain: Who is an impact investor? What are returns? And yet the things that matter the most happen at the other end of the value chain, at the level of customers and the enterprises that serve them.

We must continue to push further and faster in our work to analyze how people’s lives are or are not improving as a result of our work. This begins with the simple expectation that one cannot be an impact investor if one cannot articulate the impact one aims to have and assess whether or not that impact occurred. The IRIS taxonomy and the GIIRS ratings system serve as strong foundations for these efforts, but they are just a starting point. We need to continue to push for better, more cost-effective forms of data collection—including integrating technology into our measurement efforts—to learn more about who customers are, what their needs are, how those needs are or are not being met, and, at a minimum, what outcomes occurred.

Without this type of information we risk creating a Potemkin village, one that looks nice from the outside but crashes to the ground the moment you knock on the door to peek inside.